Open Access

ARTICLE

Evolutionary Algorithms Applied to Estimation of Thermal Property by Inverse Problem

V.C. Mariani¹, V. J. Neckel², L. S. Coelho³

1 Departamento de Engenharia Mecânica, Pontifícia Universidade Católica do Paraná - PUCPR, Curitiba, PR, Brasil. Departamento de Engenharia Elétrica, Universidade Federal do Paraná –UFPR, Curitiba, PR, Brasil {viviana.mariani@pucpr.br}
2 Departamento de Engenharia Mecânica, Pontifícia Universidade Católica do Paraná - PUCPR, Curitiba, PR, Brasil. {vagnerneckel@yahoo.com.br}
3 Departamento de Engenharia Mecatrônica, Pontifícia Universidade Católica do Paraná – PUCPR, Curitiba, PR, Brasil. Departamento de Engenharia Elétrica, Universidade Federal do Parané –UFPR, Curitiba, PR, Brasil {leandro.coelho@pucpr.br}

Computer Modeling in Engineering & Sciences 2010, 68(2), 167-184. https://doi.org/10.3970/cmes.2010.068.167

Abstract

In this study an inverse heat conduction problem using two optimization methods to estimate apparent thermal diffusivity at different drying temperatures is solved. Temperature and moisture versus time were obtained numerically using heat and mass transfer equations with drying temperatures in the range between 20°C to 70°C. The solution of the partial differential equation is made with a finite difference method coupled to optimization techniques of Differential Evolution (DE) and Particle Swarm Optimization (PSO) used in inverse problem. Statistical analysis shows no significant differences between reported and estimated curves, and no remarkable differences between results obtained using DE and PSO in 30 runs. The convective and evaporative effects and shrinkage assumptions in the model provides greater reliability on the calculated thermal diffusivity.

---

Keywords

---